ORMS in the Service Sector

1
OR/MS in the Service Sector

• Ronald L (Ron) Rardin
• School of Industrial Engineering
• Purdue University
• West Lafayette, IN 47907

2
The Service Sector

• Service enterprises dominate the U.S. economy
• 70-80 of employment
• 60-70 of GDP
• Health care alone is 14-15 of GDP
• Provide key infrastructure (e.g. transportation)

(Source Herzenberg, Alic and Wial, 1998)
3
The Service Sector

• Quality and productivity gains in manufacturing
  (some NSF inspired) have not been matched in
  services
• 4.5 productivity growth per year in
  manufacturing through the 1990s vs. 0.5 for
  service
• Services suffer a general lack of respect
• Like the ... tune of the musician, a service
  perishes in the very instant of its production,
  and seldom leaves any trace or value behind
  (Adam Smith 1776)

4
Difference Intangibility

• By definition, service are intangible
• Cannot inventory to buffer uncertainty and smooth
  production
• Consequence service operations much more demand
  responsive
• getting productive resources where/when customers
  demand

5
Intangible Position/Schedule

• Issue positioning equipment where needed
• airplanes, rail cars, rent cars, truck trailers,
  hospital respirators
• Issue scheduling staff to meet demand 24/7
• services rarely work fixed shifts like
  manufacturing
• many experiments add temp, home, or overseas
  workers

6
Intangible Demand Mgmt

• Issue dynamic pricing price according to
  demand
• widely used in airlines, hotels, rental cars,
  tourism spreading
• Issue appointments/reservations
• why make MD appointment months ahead and still
  wait hours?
• how much should airlines, hotels, restaurants,
  MDs overbook?

7
Old/New Retail Layout

• Long research on facilities layout in mfg and
  distribution
• Task arrange spaces and aisles to minimize
  materials handling traffic/cost
• In retail the task is still arranging spaces
  aisles
• New Objective maximize customer time, especially
  at high profit items

8
Old/New Medical Errors

• Wrong medication, test, etc.
• 80-100K deaths, 8-9B per year
• Much is informatics
• e.g. bar codes, MD order entry
• Adverse event monitoring raises new issues in
  stat process control
• non-homogeneous because of patient condition
  (e.g. surgery)
• auto-correlated through time (e.g. diabetes)

9
Old/New IT-Based Routing

• Vehicle routing is an old prob
• harder if both pickup drop
• harder if time windows
• Modern communication, GPS and new modes raise new
  issues
• shift stop assignments?
• accept new pickup stops?
• how to support internet-based delivery from
  several pickups?

10
Old/New Homeland Security

• Issue Re-engineering for resilience in the face
  of catastrophic events
• how much redundancy is needed and where?
• how can greater flexibility be created without
  adding excessive cost?

11
Old/New Homeland Security

• Issue Introducing security without destroying
  efficiency service
• deterrence vs. detection?
• where to place detection devices how many?
• what fraction to inspect?
• impact of preferred customer bypass systems?

12
Lots More Services Issues

• Financial engineering of markets, auctions,
  investments and tools for risk management
• Health care network design
• Outsourcing related enterprise design issues in
  business services
• Collab forecasting marketing in retail
• Payment incentive models in health care
• Transportation fleet management
• Call center training assignment